Re: improve DEBUG1 logging of parallel workers for CREATE INDEX? - Mailing list pgsql-hackers

From Tom Lane
Subject Re: improve DEBUG1 logging of parallel workers for CREATE INDEX?
Date
Msg-id 1692530.1736369905@sss.pgh.pa.us
Whole thread Raw
In response to Re: improve DEBUG1 logging of parallel workers for CREATE INDEX?  (Sami Imseih <samimseih@gmail.com>)
Responses Re: improve DEBUG1 logging of parallel workers for CREATE INDEX?
List pgsql-hackers
Sami Imseih <samimseih@gmail.com> writes:
> I am curious what are the thoughts on introducing a
> CREATE INDEX VERBOSE which can provide this info?
> similar to users scripting VACUUM VERBOSE to log
> more details about the vacuum operation including parallel
> usage.

What I can recall being discussed in the past is to extend
EXPLAIN and/or EXPLAIN ANALYZE to cover utility statements
that have nontrivial execution complexity --- for example,
ALTER TABLE has a lot of machinery underneath, and people
often wish to know things like whether a particular ALTER
will cause a table rewrite or not.

Of course, a patch for that would be a few orders of magnitude
larger than what you've got here :-(.  But if you're looking
for a framework for reporting these sorts of details, I'd
much rather go in that direction than follow the model of
VACUUM VERBOSE.  VACUUM VERBOSE is a kluge with little to
recommend it other than having been easy to implement.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Guillaume Lelarge
Date:
Subject: Re: improve DEBUG1 logging of parallel workers for CREATE INDEX?
Next
From: Andres Freund
Date:
Subject: Re: AIO v2.0